Why Both Bulls and Bears Are Often So Bird-Brained

By JONATHAN FUERBRINGER

arkets must be rational: most distinguished economists have bet their careers on that.

Still, your gut may be telling you something quite different -- that investors can be just plain irrational.

You are not alone.

To measure just how quirky investors can be, some insurgent economists are going beyond mathematical equations and computer models and are engaging in some unusual behavior of their own. They are joining psychologists in observing how laboratory humans perform in all sorts of experiments. They have watched normally rational citizens flip coins, to test their inclination to gamble. They are consulting old studies of pigeons to learn how people can deceive themselves.

And they are asking: Why do investors do things that are so obviously harmful? Why do they see patterns where there are none? Why do they behave with such overconfidence when common sense says otherwise?

The answers, these economists believe, can be crucial to understanding the stock market at a time when the chairman of the Federal Reserve shakes the financial world with the frightening words "irrational exuberance."

The occasional irrational binges of investors, they argue, have helped explain why the traditional theory of market performance periodically misfires, although last Thursday's 140-point drop in the Dow Jones industrial average may be no more than a rational response to the Fed's increase in interest rates.

More practically, individual investors and money managers can benefit enormously by recognizing their irrational alter egos: the otherwise sensible investor who is prone to excessive, costly trading; who ignores history and is lured by the market action of the moment; who can't stand to take losses but ultimately is humbled into suffering yet bigger losses; whose worst fear is investor's remorse.

"If you know there are fairly consistent patterns of behavior bias and can go against those patterns, you can make money," said Susan Belden, the co-editor of the No-Load Fund Analyst, a money-management newsletter based near San Francisco.

Value investing, the investing style most dependent on behavioral quirks, has done this for a long time, capitalizing on the tendency of people to overreact to bad news and to react slowly to good news. This gives the investor a chance to profit by buying stocks sold off by the herd.

"Fundamentally, this approach tells us that human behavior matters in markets just like everywhere else," said Richard H. Thaler, a professor of economics at the University of Chicago and a leader in the field of behavioral economics.

N. Gregory Mankiw, a more traditional professor of economics at Harvard and the author of a new introductory economics textbook, agrees. "Maybe to fully understand market behavior we have to introduce some form of irrationality," he acknowledged. The behaviorists have "opened my eyes to alternative ways of thinking."

Financial behaviorists, who are being heeded as never before, are building on years of research by psychologists and others with no economic axes to grind, adding their own experiments and new conclusions. But their interpretations of behavior and how it applies to financial markets are less precise than the equations and computer models of traditional economists.

The traditional theory of markets holds that investors are rational and that markets are efficient, quickly reflecting new information and the decisions of millions of investors.

Based on this view, investors should hold stocks for the long term, unless new information dictates a change. After all, history tells us that investing for the long run is the surest route to profit, and that with information disseminated so quickly and then promptly reflected in stock prices, we cannot act fast enough to outperform the market.

Still, investors can exhibit persistent and potentially harmful symptoms of overconfidence, unshaken in their convictions that they can beat the market. Both professional and amateur investors continue to trade actively, suffering needless losses and costing themselves a bundle in commissions. And professional money managers are thriving because of investors' enduring hopes that the stock pickers ultimately will excel.

This mentality has been tested in surveys of car drivers. Asked it they consider themselves above-average drivers, most people say yes, leaving wide open the question of who all the below-average drivers are.

"One of the hardest things to imagine is that you are not smarter than average," said Daniel Kahneman, a professor of psychology and public affairs at Princeton University.

Thaler and Robert J. Shiller, an economics professor at Yale, note that individual investors and money managers persist in their beliefs that they are endowed with more and better information than others, and that they can profit by picking stocks.

Sobering experience can help those who delude themselves. But not always. That people "do not learn to correct most of their tendencies to overconfidence is apparently just one of the limitations of the human mind," Shiller wrote in an article available on the World Wide Web.

Overconfident investors, behaviorists say, also affect the market as a whole. Consider the huge trading volume in stocks. Much of it, some economists believe, can be traced to the behaviorial quirks of investors.

High volume is "the single most embarrassing fact for an efficient market," Thaler said. "People are supposed to be buying and holding."

Terrance Odean, a behavioral economist who will join the faculty of the University of California at Davis in May, found a stunning pace of buying and selling in his study of trading at an unidentified discount brokerage firm from 1987 to 1993. Examining 10,000 accounts, Odean found that individual investors had an average annual turnover rate of 78 percent in the securities they owned.

By his calculations, the stocks that they sold outperformed the ones they bought by about 3 percentage points, before commissions, in the year after the sale,. Such a missed opportunity is the price that investors pay for overconfidence about their own investment prowess.

Could overconfidence also foster the "irrational exuberance" over stocks that has worried the chairman of the Federal Reserve? Kahneman thinks so. Investors may think they can continue doing well in the market despite warnings from many analysts that stocks are overvalued.

"Optimism has a lot to do with what is going on," he said. "Many more than 50 percent of the people in the market think they are better than average at picking stocks and picking trends, and that contributes to irrational exuberance."

If stocks are likely to produce a much higher return than bonds over the long run, why do investors still buy bonds?

Because, the behaviorists say, the pain of a small loss is greater than the thrill of a big gain. In addition, it seems that investors become even more averse to losses through apparently irrelevant actions, like frequently checking the value of their portfolios.

As Peter L. Bernstein observed in "Against the Gods," his 1996 book about the history of risk-taking, "Losses will always loom larger than gains."

Thaler argues that this fear of losses makes investors ignore the opportunity for gains by sticking with stocks. Staying out of the stock market during the last two years, he says, has probably meant more "losses," in missed opportunities, than the out-of-pocket losses from the 1987 crash.

Another behavior -- avoiding the pain of regret -- helps explain why investors hold on to losers for too long. By not selling, the investor avoids fully facing the fact that he made a bad decision.

Investors also tend to take much more risk to avoid losses than to secure gains. That behavior is demonstrated in an experiment once conducted by Kahneman and the late Amos Tversky, two leading contributors to the psychology behind behavioral finance.

In the experiment, participants were told that there was a rare disease breaking out and that 600 people were expected to die. Two plans were proposed. Under Plan A, 200 people would survive. Under Plan B, there was a 33 percent chance that everyone would be saved and a 67 percent probability that no one would be saved. Seventy-two percent of the people chose the risk-averse Plan A, with the certainty of saving 200 lives.

Then the same participants were offered another set of alternatives. Under Plan C, 400 people would die. Under Plan D, there was a 33 percent probability that no one would die and a 67 percent chance that everyone would die.

Plan A and C have the same outcome (200 people survive), as do Plans B and D (a 33 percent chance that all would survive but a 67 percent chance that no one would). But the way the second set of alternatives was phrased highlighted the loss of life -- the certainty that 400 people would die. And in that circumstance, 78 percent chose to gamble on the 33 percent chance that no one would die.

This discomfort with losses may help explain why some investors buy more shares of their losing stocks when prices fall further. They don't want to confront the loss, so they take added risk -- by buying more of the stock -- in an attempt to avoid the loss.

According to the behaviorists, investors can suffer from "magical thinking." They make connections between one occurrence and a subsequent one when, in fact, there is no link.

In illustrating such self-deception, behaviorists mention an old study of pigeons. In the 1940s, B.F. Skinner, the psychologist, fed starved, caged pigeons small amounts of food at 15-second intervals, regardless of the birds' behavior. The birds began to behave as if the feeding was a response to their behavior at the time.

One pigeon that happened to have been doing turns at feeding time began its turns again at subsequent feeding times. Another pigeon that had been bobbing its head would repeat its behavior at the next feeding time, expecting more food.

"A lot of what goes on in the markets is no more explainable than those pigeons," Shiller argued, highlighting the role of behavior in financial markets.

In the article on his Web home page, Shiller suggested that magical thinking might explain why the stock market has tended to fall in recent years on positive news about economic growth.

The news media, he wrote, have frequently attributed that seemingly perverse reaction to expectations that stronger economic growth would force the Federal Reserve to raise interest rates, which would be bad news for stocks.

In fact, Shiller concluded, an entirely random series of dips in the stock market after positive economic news might have convinced the media and investors that there was a connection between the events when none existed.

"The whole belief could be the result of a chain of events that was set off by some initial chance movements of the stock market," Shiller wrote. "Because people believe these theories, they may then behave so that the stock price does indeed behave as hypothesized."

How can two individuals make entirely different decisions when their circumstances are essentially the same?

Chalk it up to something called mental accounting, which allows individuals to treat similar results or similar options differently, because of different starting points.

In "Against the Gods," Bernstein cites an experiment in which Kahneman and Tversky asked the participants to imagine they were theatergoers headed to a Broadway show.

One group has $40 tickets. Upon arriving at the box office, however, they discover that they have lost their tickets. Distressed, most of them decide not to spend another $40 for a replacement ticket, and go home.

Another group arrives at the box office, intending to buy tickets. But they discover that $40 is missing from each of their wallets. Like the first theatergoers, they are each $40 poorer. But most of them still buy tickets and attend the show. Although both groups arrived at the theater short the same amount of money, they have mentally accounted for it differently.

Mental accountng is a way for people to organize their experiences, Kahneman says. But he adds that "it certainly can be used for self-deception." That meand investors do not always respond to events in the expected rational way.

An experiment with coin tosses conducted by Thaler shows how mental accounting can influence an investor's willingness to gamble.

In one case, each person in a group of students was given $30 and told he could walk away. Or, he could take his chances on a coin toss, winning $9 more if the coin came up heads but losing $9 if it turned up tails. Seventy percent took the gamble, knowing that they would end up with at least $21.

Students in another group were offered seemingly different gamble. They could flip a coin; heads would equal $39 and tails would equal $21. Or, they could just take $30 with no coin toss. Only 43 percent chose the gamble.

A little arithmetic shows that the two outcomes are identical. But the students made different choices. Was one group more irrational than the other?

Not necessarily. As behaviorists note, it is not just the information that determines behavior, but also (italics)how(end italics) it is processed. That is mental accounting.

Thaler says mental accounting may help explain why the 1987 market crash did not have the negative impact on the economy that many had forecast. Investors, he said, may not have seen it as a troubling actual loss of wealth, but instead viewed it less alarmingly as the disappearance of a new-found gain. And so, no panic.

The economic extablishment has come a long way in understanding behavioral about faces and unexpected responses.

Twenty years ago, when he was a student at Princeton, Mankiw of Harvard could not have taken an economics course in irrational behavior. None existed. Now it has become respectable for professors at Harvard, Yale and the University of Chicago to pick holes in the traditional view that investors are rational and markets are efficient.

The National Bureau of Economic Research and the John F. Kennedy School of Government at Harvard will hold separate symposiums in April on behavior and its influence on financial markets; the first meeting is for academics, the second for money managers.

Last fall, U.S. Trust, a money management firm for wealthy individuals, asked two of the behavioral thinkers, Kahneman and Thaler to brief its money managers.

And there are even grudging but respectful nods to behaviorists from the high priests of rational market behavior. In his 1996 edition of "A Random Walk Down Wall Street," Burton Malkiel, a Princeton economics professor, noted the words of Yale's Shiller, who has counseled that "one must look to behavioral considerations and to crowd psychology to explain the actual process of price determination in the stock market."

Shiller's encounter some months back with Alan Greenspan, the Federal Reserve chairman, has given behaviorists yet another little lift. Shiller John Campbell, a professor of economics at Harvard, were invited to join three top Wall Street equity strategists for a meeting with Greenspan last December.

Two days after that session, the Fed chairman, in a speech, posed his now-famous question of whether "irrational exuberance" had pushed the stock market too high.

Shiller did not provide Greenspan with his explosive phrase; the chairman had already inserted it into earlier drafts of his speech. But that didn't mean they weren't on the same wavelength. "I am pretty sure that I used the word 'irrational,' " Shiller said, recalling the December meeting. "But not 'exuberance.' "

Later, in congressional testimony in February, Greenspan pinpointed who might be irrational. "It is not markets that are irrational -- markets merely reflect the average, the values of people," he said. "It's people who become irrationally exuberant on occasion and take actions that induce what economists like to call bubbles, which eventually burst."

Despite his choice of language, Greenspan apparently remains a believer in the essential efficiency and rationality of markets. And why not? Mainstream economic theory has done a workmanlike job in describing how markets function.

And though the traditional economists cannot explain certain anomalies -- the curious detours from the norm that are so intriguing -- the behavioral crowd has yet to come up with an overarching theory to explain how markets function.

Behaviorists "have pointed out the defects in the existing model but have not come up with a better model," Mankiw said.

Nor does the sniping of behaviorists mean that markets over all, and over the long run, are irrational.

"To say that investors are affected by behavioral motivation does not necessarily mean that markets are not efficient," said Odean, once an undergraduate student of Kahneman.

But that does not deny the application to individual investors, who, if they are honest with themselves, know that they do not always behave rationally.

Dr. Michael Apfel, 47, an oral surgeon from Phoenix, acknowledges that he could be viewed as a case study. He freely confesses that behavioral quirks have undermined his investment performance.

A look at his portfolio suggests that he is the quintessential conservative investor. He dutifully balanced his portfolio with stocks, bonds, mutual funds, savings bonds and real estate partnerships. And yet, he said, "I used to buy and sell quite regularly." The cost of trading alone had cut his gains by 25 percent.

He had trouble getting rid of losing stocks, too. "One doesn't want to admit defeat," he said.

Now he trades only with the idea of making changes in his long-term investment posture. And he jettisons losers as part of a deliberate tax strategy to offset capital gains.

"You have to be there all the time to do well," he said of active trading. "Or you have to be a long-term investor. Anywhere in between and you are a loser."

What he has learned is that investors are better off if they know how behavior can trip them up. But millions of other investors have yet to share Apfel's epiphany. Perhaps if they were fully rational, they would have by now.